scholarly journals Urban vegetation classification with high-resolution PlanetScope and SkySat multispectral imagery

2021 ◽  
Vol 15 (1) ◽  
pp. 66-75
Author(s):  
Loránd Szabó ◽  
Dávid Abriha ◽  
Kwanele Phinzi ◽  
Szilárd Szabó
Keyword(s):  
High Resolution ◽  
Input Data ◽  
Roc Curves ◽  
Classification Performance ◽  
Support Vector ◽  
Urban Vegetation ◽  
Classification Methods ◽  
Linear Modeling ◽  
Spectral Bands ◽  
The Difference

In this study two high-resolution satellite imagery, the PlanetScope, and SkySat were compared based on their classification capabilities of urban vegetation. During the research, we applied Random Forest and Support Vector Machine classification methods at a study area, center of Rome, Italy. We performed the classifications based on the spectral bands, then we involved the NDVI index, too. We evaluated the classification performance of the classifiers using different sets of input data with ROC curves and AUC values. Additional statistical analyses were applied to reveal the correlation structure of the satellite bands and the NDVI and General Linear Modeling to evaluate the AUC of different models. Although different classification methods did not result in significantly differing outcomes (AUC values between 0.96 and 0.99), SVM’s performance was better. The contribution of NDVI resulted in significantly higher AUC values. SkySat’s bands provided slightly better input data related to PlanetScope but the difference was minimal (~3%); accordingly, both satellites ensured excellent classification results.

scholarly journals Melamine Faced Panels Defect Classification beyond the Visible Spectrum

Sensors ◽  
2018 ◽  
Vol 18 (11) ◽  
pp. 3644
Author(s):  
Cristhian Aguilera ◽  
Cristhian Aguilera ◽  
Angel Sappa
Keyword(s):  
Near Infrared ◽  
Visible Spectrum ◽  
Local Binary Patterns ◽  
Classification Performance ◽  
Support Vector ◽  
Feature Descriptor ◽  
Long Wavelength ◽  
Spectral Bands ◽  
Image Set ◽  
Long Wavelength Infrared

In this work, we explore the use of images from different spectral bands to classify defects in melamine faced panels, which could appear through the production process. Through experimental evaluation, we evaluate the use of images from the visible (VS), near-infrared (NIR), and long wavelength infrared (LWIR), to classify the defects using a feature descriptor learning approach together with a support vector machine classifier. Two descriptors were evaluated, Extended Local Binary Patterns (E-LBP) and SURF using a Bag of Words (BoW) representation. The evaluation was carried on with an image set obtained during this work, which contained five different defect categories that currently occurs in the industry. Results show that using images from beyond the visual spectrum helps to improve classification performance in contrast with a single visible spectrum solution.

scholarly journals Explaining Machine Learning Models for Clinical Gait Analysis

2022 ◽  
Vol 3 (2) ◽  
pp. 1-27
Author(s):  
Djordje Slijepcevic ◽  
Fabian Horst ◽  
Sebastian Lapuschkin ◽  
Brian Horsak ◽  
Anna-Maria Raberger ◽  
...  
Keyword(s):  
Machine Learning ◽  
Reaction Force ◽  
Statistical Parametric Mapping ◽  
Qualitative Evaluation ◽  
Classification Performance ◽  
Healthcare Sector ◽  
Support Vector ◽  
Classification Methods ◽  
Network Support ◽  
Gait Classification

Machine Learning (ML) is increasingly used to support decision-making in the healthcare sector. While ML approaches provide promising results with regard to their classification performance, most share a central limitation, their black-box character. This article investigates the usefulness of Explainable Artificial Intelligence (XAI) methods to increase transparency in automated clinical gait classification based on time series. For this purpose, predictions of state-of-the-art classification methods are explained with a XAI method called Layer-wise Relevance Propagation (LRP). Our main contribution is an approach that explains class-specific characteristics learned by ML models that are trained for gait classification. We investigate several gait classification tasks and employ different classification methods, i.e., Convolutional Neural Network, Support Vector Machine, and Multi-layer Perceptron. We propose to evaluate the obtained explanations with two complementary approaches: a statistical analysis of the underlying data using Statistical Parametric Mapping and a qualitative evaluation by two clinical experts. A gait dataset comprising ground reaction force measurements from 132 patients with different lower-body gait disorders and 62 healthy controls is utilized. Our experiments show that explanations obtained by LRP exhibit promising statistical properties concerning inter-class discriminativity and are also in line with clinically relevant biomechanical gait characteristics.

scholarly journals Pengenalan Pola Tulisan Tangan Suku Kata Aksara Sasak Menggunakan Metode Moment Invariant dan Support Vector Machine

2019 ◽  
Vol 3 (2) ◽  
Author(s):  
Riska Yulianti ◽  
I Gede Pasek Suta Wijaya ◽  
Fitri Bimantoro
Keyword(s):  
Support Vector Machine ◽  
Input Data ◽  
Performance Indicator ◽  
Classification Performance ◽  
Support Vector ◽  
Image Transformation ◽  
Moment Invariant ◽  
Image Scaling ◽  
Performance Accuracy ◽  
Average Accuracy

 The research of Javanese and Balinese ancient script have been done by some researches. However, the researches still have problems, such as image scaling, noise reduction and image transformation. This research implemented moment invariant and support vector machine to solve these problems especially on Sasak ancient script. The input data used in this research was 2700 handwritten Sasak ancient script. The testing was done to know the effect of thinning and the number of feature by using zoning on the classification performance. Accuracy is used as performance indicator. The highest average accuracy is 89.76%, on the second scenario, the average accuracy obtained is 92.52%. 

scholarly journals Classification of Hyperspectral Images with CNN in Agricultural Lands

2021 ◽  
Vol 3 (1) ◽  
pp. 6
Author(s):  
Eren Can Seyrek ◽  
Murat Uysal
Keyword(s):  
Classification Performance ◽  
Hyperspectral Images ◽  
Superior Performance ◽  
Support Vector ◽  
World Population ◽  
Agricultural Lands ◽  
Leading Role ◽  
Spectral Bands ◽  
Land Cover Maps

Hyperspectral images (HSI) offer detailed spectral reflectance information about sensed objects through provision of information on hundreds of narrow spectral bands. HSI have a leading role in a broad range of applications, such as in forestry, agriculture, geology, and environmental sciences. The monitoring and management of agricultural lands is of great importance for meeting the nutritional and other needs of a rapidly and continuously increasing world population. In relation to this, classification of HSI is an effective way for creating land use and land cover maps quickly and accurately. In recent years, classification of HSI using convolutional neural networks (CNN), which is a sub-field of deep learning, has become a very popular research topic and several CNN architectures have been developed by researchers. The aim of this study was to investigate the classification performance of CNN model on agricultural HSI scenes. For this purpose, a 3D-2D CNN framework and a well-known support vector machine (SVM) model were compared using the Indian Pines and Salinas Scene datasets that contain crop and mixed vegetation classes. As a result of this study, it was confirmed that use of 3D-2D CNN offers superior performance for classifying agricultural HSI datasets.

Supervised Learning for Binary Classification on US Adult Income

2021 ◽  
Vol 13 (2) ◽  
pp. 80-91
Author(s):  
Li-Pang Chen
Keyword(s):  
Missing Values ◽  
Binary Classification ◽  
Roc Curves ◽  
Income Level ◽  
Support Vector ◽  
Classification Methods ◽  
Gender Education ◽  
The Us ◽  
Auc Value ◽  
R Functions

In this project, various binary classification methods have been used to make predictions about US adult income level in relation to social factors including age, gender, education, and marital status. We first explore descriptive statistics for the dataset and deal with missing values. After that, we examine some widely used classification methods, including logistic regression, discriminant analysis, support vector machine, random forest, and boosting. Meanwhile, we also provide suitable R functions to demonstrate applications. Various metrics such as ROC curves, accuracy, recall and F-measure are calculated to compare the performance of these models. We find the boosting is the best method in our data analysis due to its highest AUC value and the highest prediction accuracy. In addition, among all predictor variables, we also find three variables that have the largest impact on the US adult income level.

scholarly journals OBJECT-BASED GREENHOUSE CLASSIFICATION FROM HIGH RESOLUTION SATELLITE IMAGERY: A CASE STUDY ANTALYA-TURKEY

2016 ◽  
Vol XLI-B7 ◽  
pp. 183-187 ◽  
Author(s):  
M. Coslu ◽  
N. K. Sonmez ◽  
D. Koc-San
Keyword(s):  
High Resolution ◽  
Satellite Imagery ◽  
Accuracy Assessment ◽  
Digital Data ◽  
Classification Method ◽  
Classification Methods ◽  
Spectral Bands ◽  
Object Based ◽  
Rule Sets ◽  
High Resolution Satellite Imagery

Pixel-based classification method is widely used with the purpose of detecting land use and land cover with remote sensing technology. Recently, object-based classification methods have begun to be used as well as pixel-based classification method on high resolution satellite imagery. In the studies conducted, it is indicated that object-based classification method has more successful results than other classification methods. While pixel-based classification method is performed according to the grey value of pixels, object-based classification process is executed by generating imagery segmentation and updatable rule sets. In this study, it was aimed to detect and map the greenhouses from object-based classification method by using high resolution satellite imagery. The study was carried out in the Antalya province which includes greenhouse intensively. The study consists of three main stages including segmentation, classification and accuracy assessment. At the first stage, which was segmentation, the most important part of the object-based imagery analysis; imagery segmentation was generated by using basic spectral bands of high resolution Worldview-2 satellite imagery. At the second stage, applying the nearest neighbour classifier to these generated segments classification process was executed, and a result map of the study area was generated. Finally, accuracy assessments were performed using land studies and digital data of the area. According to the research results, object-based greenhouse classification using high resolution satellite imagery had over 80% accuracy.

scholarly journals Image-Based Learning Approach Applied to Time Series Forecasting

2012 ◽  
Vol 10 (3) ◽  
Author(s):  
K. Ramírez-Amáro ◽  
J. C. Chimal-Eguía
Keyword(s):  
Time Series ◽  
Input Data ◽  
Data Representation ◽  
Time Series Forecasting ◽  
Support Vector ◽  
Learning Approach ◽  
New Information ◽  
New Learning ◽  
Learning Technique ◽  
The Difference

In this paper, a new learning approach based on time-series image information is presented. In order to implement this new learning technique, a novel time-series input data representation is also defined. This input data representation is based on information obtained by image axis division into boxes. The difference between this new input data representation and the classical is that this technique is not time-dependent. This new information is implemented in the new Image-Based Learning Approach (IBLA) and by means of a probabilistic mechanism this learning technique is applied to the interesting problem of time series forecasting. The experimental results indicate that by using the methodology proposed in this article, it is possible to obtain better results than with the classical techniques such as artificial neuronal networks and support vector machines.

Comparative Study of NMR Spectral Profiling for the Characterization and Authentication of Cannabis

2017 ◽  
Vol 100 (5) ◽  
pp. 1356-1364 ◽  
Author(s):  
Xinyi Wang ◽  
Peter de B Harrington ◽  
Steven F Baugh
Keyword(s):  
Discriminant Analysis ◽  
Nmr Spectroscopy ◽  
Plant Extracts ◽  
Screening Method ◽  
Fuzzy Rule ◽  
Classification Performance ◽  
Support Vector ◽  
Classification Methods ◽  
Linear Discriminant ◽  
Projected Difference Resolution

Abstract For the authentication of botanical materials, itis difficult to obtain representative reference materials because botanicals vary significantly with respect to cultivation conditions. Chemical profiling of plant extracts or spectral fingerprinting can differentiate botanicals and group them by their chemical profiles. NMR spectroscopy yields a powerful and useful method for profiling plant extracts. Both 500 MHz 1H and 1H-1H correlation NMR spectroscopy coupled with pattern recognition were used to discriminate among Cannabis samples. A rapid method of analysis was achieved by extracting directly into the deuterated solvent. Spectral ranges including or excluding the downfield region were compared to evaluate the effect on classification accuracy by projected difference resolution. Six classification methods—fuzzy rule-building expert system, linear discriminant analysis (LDA), super partial least-squares discriminant analysis, support vector machine (SVM), and SVMclassification trees (SVMTrees)—all gave better classification performance for proton NMR spectrathan for proton-proton correlation NMR spectra for seven Cannabis samples. Among the classification methods for a set of 25 Cannabis samples, the 0.5–7.2 plus 7.4–13.0 ppm ranges gave higher prediction rates of greater than 96% when compared to the reduced range of 0.5–7.2 ppm that excluded the downfield range. The LDA method had the best prediction accuracy of 99.8 ± 0.4%. SVMTree methods provide a robust tool, and classification trees are amenable to interpretation. Hence, NMR spectroscopy combined withchemometrics could be used as a fast screening method for the authentication of Cannabis samples.

scholarly journals Classification Performance for Making Decisions about Products Missing from the Shelf

2011 ◽  
Vol 2011 ◽  
pp. 1-13 ◽  
Author(s):  
Dimitris Papakiriakopoulos ◽  
Georgios Doukidis
Keyword(s):  
Area Under Curve ◽  
Naive Bayes ◽  
Predictive Accuracy ◽  
Roc Curves ◽  
Classification Performance ◽  
Retail Store ◽  
Classification Algorithms ◽  
Classification Methods ◽  
Feature Spaces ◽  
Retail Chains

The out-of-shelf problem is among the most important retail problems. This work employs two different classification algorithms, C4.5 and naïve Bayes, in order to build a mechanism that makes decisions about whether a product is available on a retail store shelf or not. Following the same classification methods and feature spaces, we examined the classification performance of the algorithms in four different retail chains and utilized ROC curves and the area under curve measure to compare the predictive accuracy. Based on the results obtained for the different retail chains, we identified certain approaches for the development and introduction of such a mechanism in different retail contexts.

scholarly journals A Hybrid Feature Selection and Extraction Methods for Sleep Apnea Detection Using Bio-Signals

Sensors ◽  
2020 ◽  
Vol 20 (15) ◽  
pp. 4323 ◽  
Author(s):  
Xilin Li ◽  
Sai Ho Ling ◽  
Steven Su
Keyword(s):  
Feature Selection ◽  
Sleep Apnea ◽  
Regularization Parameter ◽  
Roc Curves ◽  
Extraction Methods ◽  
Classification Performance ◽  
Hill Climbing ◽  
Support Vector ◽  
Svm Classifier ◽  
Increased Risk

People with sleep apnea (SA) are at increased risk of having stroke and cardiovascular diseases. Polysomnography (PSG) is used to detect SA. This paper conducts feature selection from PSG signals and uses a support vector machine (SVM) to detect SA. To analyze SA, the Physionet Apnea Database was used to obtain various features. Electrocardiography (ECG), oxygen saturation (SaO2), airflow, abdominal, and thoracic signals were used to provide various frequency-, time-domain and non-linear features (n = 87). To analyse the significance of these features, firstly, two evaluation measures, the rank-sum method and the analysis of variance (ANOVA) were used to evaluate the significance of the features. These features were then classified according to their significance. Finally, different class feature sets were presented as inputs for an SVM classifier to detect the onset of SA. The hill-climbing feature selection algorithm and the k-fold cross-validation method were applied to evaluate each classification performance. Through the experiments, we discovered that the best feature set (including the top-five significant features) obtained the best classification performance. Furthermore, we plotted receiver operating characteristic (ROC) curves to examine the performance of the SVM, and the results showed the SVM with Linear kernel (regularization parameter = 1) outperformed other classifiers (area under curve = 95.23%, sensitivity = 94.29%, specificity = 96.17%). The results confirm that feature subsets based on multiple bio-signals have the potential to identify patients with SA. The use of a smaller subset avoids dimensionality problems and reduces the computational load.

Sign in / Sign up

Export Citation Format

Share Document